GLCM features and fuzzy nearest neighbor classifier for emotion recognition from face

An emotion recognition method from the face images is proposed in this paper, which can recognize seven emotions of human, i.e., six basic expressions in addition to neutral. The proposed method uses the GLCM approach for feature extraction and the nearest neighbor (NN) for classification. The fuzzy Euclidean distance is used. GLCM provides the texture characteristics of an input image through the second order statistical measurements. Because of existence of vagueness and uncertainty in the discriminant features extracted from different emotional face images, the fuzzy measure is involved in the NN classifier to recognize the emotions of faces with more accuracy. The experiments show the good efficiency of the introduced recognition method compared to some other feature extraction and facial emotion recognition methods.